-
Development of sustainable and functionalized
inorganicbinder-biofiber compositesDoudart de la Gre, G.C.H.
Accepted/In press: 17/09/2018
Document VersionPublishers PDF, also known as Version of Record
(includes final page, issue and volume numbers)
Please check the document version of this publication:
A submitted manuscript is the author's version of the article
upon submission and before peer-review. There can be important
differencesbetween the submitted version and the official published
version of record. People interested in the research are advised to
contact theauthor for the final version of the publication, or
visit the DOI to the publisher's website. The final author version
and the galley proof are versions of the publication after peer
review. The final published version features the final layout of
the paper including the volume, issue and page numbers.
Link to publication
General rightsCopyright and moral rights for the publications
made accessible in the public portal are retained by the authors
and/or other copyright ownersand it is a condition of accessing
publications that users recognise and abide by the legal
requirements associated with these rights.
Users may download and print one copy of any publication from
the public portal for the purpose of private study or research. You
may not further distribute the material or use it for any
profit-making activity or commercial gain You may freely distribute
the URL identifying the publication in the public portal ?
Take down policyIf you believe that this document breaches
copyright please contact us providing details, and we will remove
access to the work immediatelyand investigate your claim.
Download date: 31. Aug. 2018
https://research.tue.nl/en/publications/development-of-sustainable-and-functionalized-inorganic-binderbiofiber-composites(79a5e2e2-8232-4bbe-8ddb-83d4b3a771bc).html
-
/ Department of the Built Environment/ Department of the Built
Environment
This thesis addresses the performance-based design and
evaluation of lig-nocellulosic cement composite boards, which are
commercially known as wood wool cement boards (WWCB). The origin of
WWCB goes back to around 1920 and is still popular nowadays. This
thesis starts with an introduction of WWCB by explaining its
ingredients and the production process. Next, the retardation of
sugars on cements is evalu-ated, providing new insights into the
retardation mechanism when combining wood and cement. To reduce the
environmental footprint of the boards, imple-mentation of
supplementary materials and the use of alkali-activated binders are
studied. An orientated study is then performed on increasing the
functionality of the boards by making it air purifying using the
fundamental insights in the surface morphology. Since one of the
main WWCB properties is sound absorp-tion, a study is performed to
characterize and to, predict the sound absorption of WWCB by using
impedance models. Finally, the disposal of commercial WWCB after
its service life time is considered, leading to waste wood
incineration and contamination of residues that require the design
of economical feasible treat-ments.
BOUWSTENEN Proefschrift
matte laminate
book size 170 x 245 mm
Guillaume Doudart de la Gre
De promotie zal plaatsvinden in de Senaats-
zaal van het Auditorium
van de Technische Universiteit
Eindhoven
Aansluitend aan deze plechtigheid zal een
receptie plaatsvindenwaarvoor u van harte bent
uitgenodigd
Uitnodigingtot het bijwonen
van de openbare verdediging van mijn proefschrift
Development of sustai-nable and functionalized inorganic
binder-biofiber
composites
op maandag 17 September 2018
om 11.00 uur
bouwstenen 246Guillaume Doudart de la Gre
Development of sustainable and functionalized inorganic
binder-biofiber composites
Development of sustainable and functionalized inorga-
nic binder-biofiber composites
246
Spine width choice of paper duplex printed on:90 grams HVO WIT #
pages : 100 x 5.95 = spine width in mm 90 grams BIOTOP # pages :
100 x 6.40 = spine width in mm 90 grams G-PRINT silk # pages : 100
x 4.70 = spine width in mm
-
Development of sustainable and functionalized inorganic
binder-biofiber composites
G.C.H. Doudart de la Gre
-
De promotiecommissie is als volgt samengesteld:
Voorzitter:
prof.ir. E.S.M. Nelissen Technische Universiteit Eindhoven
Promotor:
prof.dr.ir. H.J.H. Brouwers Technische Universiteit
Eindhoven
Copromotor:
dr. Q.L. Yu Technische Universiteit Eindhoven
Leden (in alfabetische volgorde):
prof.dr.ing. S Amziane Universit Blaise Pascal
prof.dr.ir. N. De Belie Universiteit Gent
dr. J.E.G. van Dam Wageningen University and Research
dr.ir. M.C.J. Hornikx Technische Universiteit Eindhoven
prof.dr.ir. J.J.N. Lichtenberg Technische Universiteit
Eindhoven
Title
Development of sustainable and functionalized inorganic
binder-biofiber composites
ISBN 978-90-386-4549-0
Bouwstenen 246
Copyright 2018 by G.C.H. Doudart de la Gre
Ph.D. Thesis, Eindhoven University of Technology, the
Netherlands
Cover design: G.C.H. Doudart de la Gre
Printed by: ProefschriftMaken || www.proefschriftmaken.nl,
Vianen, the Netherlands.
All rights reserved. No part of this publication may be
reproduced in any form or by any
means without permission in writing form from the author.
A catalogue record is available from the Eindhoven University of
Technology Library.
-
Development of sustainable and functionalized inorganic
binder-biofiber composites
PROEFSCHRIFT
ter verkrijging van de graad van doctor aan de
Technische Universiteit Eindhoven, op gezag van de
rector magnificus, prof.dr.ir. F.P.T. Baaijens, voor een
commissie aangewezen door het College voor
Promoties in het openbaar te verdedigen
op maandag 17 September 2018 om 11.00 uur
door
Guillaume Claude Hendrikus Doudart de la Gre
geboren te Helmond
-
Dit proefschrift is goedgekeurd door de promotor:
prof.dr.ir. H.J.H. Brouwers
-
Dedicated to my brother Jeremy Doudart de la Gre
(April 26, 1977 June 17, 1997)
-
ii
Preface
My PhD journey started on the first of October 2012. At that
time my main goal was to
broaden my knowledge in building materials and implement the
gained knowledge in
development of improved wood cement composites. Interestingly,
performing research to
fulfill my goals was only a part of the journey and my PhD
involved much more than I
ever could have imagined. Presenting science, writing articles,
working in an innovative
environment, traveling internationally and meeting people with
cultures from all around
the world made this period of my life very special. Therefore,
this section is dedicated to
all the people that shared this journey with me providing
support and help in many
different ways.
First, I would like to express my gratitude to my promoter
prof.dr.ir. Jos Brouwers who
gave me the opportunity to join his research group and patiently
supported and
encouraged me in many phases of my PhD. Jos, thank you really
very much, you inspired
me a lot in both research and professionalism and Im grateful
that we got to know each
other so well. Next, I would like to thank my daily supervisor
dr. Qingliang Yu, for the
many guidance he provided which goes beyond research. I learned
a lot from you until
this very moment and it was a true honor to work with you on
this challenging journey.
I also appreciate the financial support from STW (Project 11861)
and of the sponsor
group who funded this research. Special thanks are given to: dr.
Jan van Dam
(Wageningen UR) for all his kind support on the lignocellulose
based knowledge during
this study; dr.ir. Arno Keulen (van Gansewinkel Minerals) for
the knowledge and
practical issues on working with industrial byproducts; Mr. Jan
Mencnarowski and Mr.
John van Eijk (Knauf Insulation) for the valuable insights
regarding the industrial
production of WWCB. Mr. Paul van Elten (Eltomation) for the
valuable knowledge of
the WWCB history and production plants for producing WWCB. Ing.
Peter de Vries
(ENCI) for the support and guidance related to implementation of
Portland Cement and
overcoming retardation issues.
I am grateful to all the members of my core committee, prof. dr.
ing. Sofiane Amziane,
prof. dr. ir. Nele De Belie, dr. Jan van Dam, dr. ir. Maarten
Hornikx, prof. dr. ir. Jos
Lichtenberg for their constructive comments on my manuscript and
for agreeing to be
members of my PhD defence committee.
During my stay at the TU/e I have met some wonderful collages
and I would like to
address my appreciation to some former and current colleagues
for their help and support:
A. de Korte, M. Florea, P. Spiesz, A. Lzaro, G. Quercia, A.
Taher, S. Lorencik, R. Yu, P.
Tang, C. Straub, P. van der Wouw, X. Gao, B. Yuan, K.
Schollbach, K. Kochov, Q.
Alam, P. Li, Z. Qu, V. Caprai, E. Loginova, G. Gauvin, A. Kaja,
H. Karimi and G. Liu.
My appreciations are also expressed to Gertjan Maas, Peter
Cappon, Anneke Delsing,
Harrie Smulders and Jan Diepens for their help in the
laboratory. My appreciations are
also expressed to our lovely secretaries of the unit BPS.
At the TU/e I had the chance to not only work on my own PhD
project but also to work
on a granted idea the leafroof and on providing several master
courses. During this
process I had the chance to work with Bert van Schaik, Henk
Schellen, Alexander
-
iii
Rosemann and Mark Cox who provided valuable feedback and
knowledge and helped me
with my professional carrier.
During my PhD I had the chance and honor of supervising two
special master students:
Veronica Caprai and Bram Botterman with whom I shared the
passion of working with
wood cement composites and who contributed to this thesis,
opening new perspectives.
Furthermore, I would like to express my gratitude to the master
students Marco de Groot,
Jonathan Ezechils and Thijs van Druenen. Thank you for your help
in assisting my
research and hopefully my supervision has been helpful for your
professional careers.
Finally, I need to express my appreciations to my supporting
family. Mum and Dad,
thank you for always being there for me and trying to understand
what I was actually
doing at the TU/e. My thanks are also given to my brother and
sister for taking care of
my son Nathan during the last years of my PhD. Finally, and most
importantly, I want to
express my appreciations and feelings to my wife, Adelya. We
started together our PhD
journey on the same date but in different BPS chairs. It was
always nice having you near
me and share our working experience. That we got married and got
Nathan in this period
was a blessing and a welcome change in life. There are not
enough words that can
express my feelings but know that I love you forever.
Guillaume Doudart de la Gre
Eindhoven, September 2018
-
iv
Contents
Preface ii
1 Introduction 8
1.1 Background 8
1.2 Raw materials 9
1.2.1 Cement
..........................................................................................................
9
1.2.2 Wood
...........................................................................................................
11
1.3 Production process WWCB in the Netherlands 16
1.3 Influential parameters and consequences 19
1.4 The influence of moisture and binder amount 23
1.5 Developments in WWCB 25
1.6 Conclusions 25
1.7 Problem statement 26
1.8 Research targets 27
1.9 Outline of this thesis 29
2 Hydration kinetics of cement containing sugars 32
2.1 Introduction 32
2.2 Materials and Methods 34
2.2.1
Materials......................................................................................................
34
2.2.1 Experiments
................................................................................................
36
2.3 Results and discussions 38
2.3.1 Hydration kinetics of cements/clinkers without glucose
............................ 38
2.3.2 Hydration kinetics of cements/clinkers containing glucose
........................ 41
2.3.3 Effect of CaSO4 content
..............................................................................
45
2.3.4 Effect of glucose
content.............................................................................
48
2.3.5 Mechanical properties of mortars and wood-cement
composites ............... 49
2.4 Conclusions 52
3 Design of alternative binders in wood composites 54
3.1 Introduction 54
3.2 Materials and methods 55
3.2.1 Characterisation of the
powders..................................................................
57
3.2.2 Mechanical properties
.................................................................................
58
3.2.3 Thermal insulating properties
.....................................................................
59
3.2.4 Sound absorbing
properties.........................................................................
59
3.2.5 Thermal analysis
.........................................................................................
59
3.3 Results and discussions 60
3.3.1 Characteristics of the binders
......................................................................
60
-
v
3.3.2 Hydration kinetics
.......................................................................................
63
3.3.3 Mechanical properties
.................................................................................
64
3.3.4 Thermal properties
......................................................................................
66
3.3.5 Acoustical
properties...................................................................................
67
3.3.6 Performance at high temperature
................................................................
68
3.4 Optimized mix design 70
3.4.1 Mix design algorithm
..................................................................................
70
3.4.2 Particle size distribution
..............................................................................
70
3.4.3 Evaluation of an optimized mix
design.......................................................
72
3.4.4 Validation
....................................................................................................
72
3.5 Conclusions 73
4 Ionic interaction and liquid absorption by wood in
lignocellulose inorganic
mineral binder composites 76
4.1 Introduction 76
4.2 Materials and Methods 78
4.2.1 Mineral binders
...........................................................................................
78
4.2.2 Lignocellulose
.............................................................................................
79
4.2.3 Characterisation of the investigated materials
............................................ 79
4.2.4 pH measurements
........................................................................................
79
4.2.5 Isothermal calorimetry measurement
.......................................................... 79
4.2.6 Ionic analyses
..............................................................................................
80
4.2.7 NMR analyses
.............................................................................................
80
4.2.8 Lignocellulose mineral binder composite
design........................................ 80
4.2.9 Environmental impact
.................................................................................
81
4.3 Results and discussion 82
4.3.1 Behaviour of wood wool under alkaline conditions
................................... 85
4.3.2 Reaction kinetics of the binders
..................................................................
87
4.3.3 Ionic uptake of wood and alteration of binder reaction
mechanism ........... 87
4.3.4 Position of the absorbed water in wood wool
strands................................. 90
4.3.5 Evaluation of pre-treatment
........................................................................
91
4.3.6 Environmental impact
.................................................................................
93
4.4 Conclusions 94
5 Advanced functionalities 96
5.1 Introduction 96
5.2 Air quality 97
5.2.1 Introduction
.................................................................................................
97
5.2.2 Indoor and outdoor air pollutants
................................................................
98
5.2.3 Additional
functionalities............................................................................
99
5.3 Photocatalytic oxidation 99
5.3.1 Introduction
.................................................................................................
99
-
vi
5.3.2 PCO working principle
.............................................................................
100
5.3.3 Application of
PCO...................................................................................
101
5.4 Air purifying WWCB 102
5.4.1 Introduction
...............................................................................................
102
5.4.2 Influential parameters
...............................................................................
102
5.4.3 Overview of the performed test
................................................................
104
5.4.4 PCO experiment
........................................................................................
107
5.5 Influential factor analysis 108
5.5.1 Introduction
...............................................................................................
108
5.5.2 Analysis of the NO degradation
................................................................
109
5.5.3 Durability
..................................................................................................
113
5.6 Conclusions 115
6 Modeling and optimization of the sound absorption of wood wool
cement
boards 118
6.1 Introduction 118
6.2 Materials and Experimental methodologies 120
6.2.1 Density and wood to binder characterisation of
WWCB.......................... 120
6.2.2 Acoustic impedance measurement
............................................................
120
6.2.3 Impedance models and their input parameters
.......................................... 120
6.3 Measured parameters and choice of impedance model 121
6.3.1 Density and wood to binder properties
..................................................... 121
6.3.2 Choice of impedance model
......................................................................
122
6.3.3 Johnson-Champoux-Allard (JCA) model
................................................. 124
6.3.4 Open porosity
............................................................................................
125
6.3.5 Flow
resistivity..........................................................................................
126
6.4 Modelling results 127
6.4.1 Analysis of the inversely calculated input parameters for
the JCA model 127
6.4.2 Validation of the impedance model
.......................................................... 129
6.5 Characterisation and optimization of the acoustical
properties of WWCB 130
6.5.1 Influence of strand width
..........................................................................
130
6.5.2 Influence of the density
.............................................................................
131
6.5.3
Optimization..............................................................................................
131
6.6 Conclusion 134
7 Disposition after service life and treatment of biomass fly
ashes 136
7.1 Introduction 136
7.2 Reuse, recycling and energy recovery 136
7.2.1 Re-use
.......................................................................................................
137
7.2.2 Recycling
..................................................................................................
137
7.2.3 Energy recovery
........................................................................................
138
7.2.4 Environmental aspects and legislation
...................................................... 140
-
vii
7.3 Characteristics of the industrial by-product 141
7.3.1
Materials....................................................................................................
141
7.3.2 Leaching
....................................................................................................
143
7.3.3 Particle size distribution
............................................................................
143
7.3.4 X-Ray Fluorescence spectroscopy
............................................................
144
7.3.5 X-Ray Diffraction
.....................................................................................
145
7.3.6 Treatment steps
.........................................................................................
146
7.4 Treatment results of biomass fly ash and a pilot washing
test 150
7.4.1 Leaching results
........................................................................................
150
7.4.2 Composition and fineness
.........................................................................
153
7.4.3 Final leaching estimation of the treated bio fly ashes
............................... 155
7.4.4 Pilot test
....................................................................................................
156
7.4.5 Economic feasibility
.................................................................................
159
7.5 Conclusions 160
8 Conclusions and recommendations 162
8.1.1 Introduction
...............................................................................................
162
8.1.2 Hydration kinetics of cement containing sugars
....................................... 162
8.1.3 Design of alternative binders for wood composites
.................................. 162
8.1.4 Ionic interaction and liquid absorption by wood in
lignocellulose inorganic
mineral binder composites
.......................................................................................
163
8.1.5 Advanced
functionalities...........................................................................
163
8.1.6 Modeling and optimization of the sound absorption of wood
wool cement
boards 164
8.1.7 Disposition after service life and treatment of biomass
fly ashes ............. 164
8.2 Recommendations for future research 165
Bibliography 166
List of symbols and abbreviations 188
Summary 192
List of publications 194
Curriculum Vitae 196
-
8 Introduction
Chapter 1
1 Introduction
1.1 Background Annually 16 million m2 of wood wool composite
boards (WWCB) are produced in
Europe. The product is known since 1900, created from spruce or
popular wood wool,
using magnesium as a binder (Wolfe, 1999; Aro, 2004). In around
1920, after the
invention of Portland Cement (PC) in the 19th century, cement
replaced the magnesium
as a binder leading to the creation of the nowadays known WWCB
(van Elten, 2006).
Because of the used cement, the term WWCB nowadays is also
translated as wood wool
cement boards. The boards became increasingly popular because of
their high thermal
insulating and sound absorbing properties gained by its high
porosity and low density. In
accordance with European Norm EN 13168 (formerly DIN 1101) the
density of WWCB
can range from approximately 360 up to 570 kg/m, although higher
and lower densities
are also possible. Consequently, because of the mineralization
of the wood wool strands,
the boards possess high resistance to bio-degradation (Pereira
et al., 2006) and fire (Aro,
2004). Hence, the boards are applied in both buildings and
constructions as roof and
ceiling material, or as an exterior wall where a high durability
and low maintenance is
required. An illustration of WWCB is presented in Figure 1.1.
The common raw
materials of WWCB are wood wool strands (lignocellulose), cement
and water (with
sometimes the addition of dissolved salts). In the northern
hemisphere mainly pine and
spruce is used, in some countries also eucalyptus. The wood is
shredded to long strands
(length 250 mm, width 1-3 mm, thickness 0.05 mm) and is dipped
in a water-
sodium silicate solution to accelerate the cement hydration. The
wetted wood, together
with cement (usually PC), is mixed and then spread on plywood
moulds and pressed.
Depending on the used distance holders, a composite of 15-50 mm
thickness is obtained.
After sufficient initial curing, e.g. after 24 hours, the
product is removed from the mould
and is left to cure further.
Figure 1.1: Illustration of a wood wool cement board with a
thickness of 15 mm.
WWCB are falling under the category wood cement composite (WCC)
and, compared to
WCC, literature regarding the product is scarce. Therefore,
information can only be
-
Chapter 1 9
derived from manufacturers or the search for information needs
to be diverted to the
general term WCC. WCC can be found under plenty of names
depending on the wood
dimensions (see an overview in Figure 1.2) referred to as
wood-strand cement board
(Aro, 2004), cement-bonded-wood particle board (Soroushian et
al., 2003), cement-
bonded composite boards (Aggarwal et al., 2008; Ashori et al.,
2011). The disadvantage
of using literature regarding WCC is the minor similarities with
WWCB since the boards
are produced in different ways, under different conditions,
using different lignocellulose
materials and face different challenges, hence, are in most
cases less applicable for
WWCB. For instance, most WCC consist of micro and macro wood
particles with a
reported inhibitory effect on the used binder resulting in lower
mechanical properties.
Moreover, the use of smaller particles also requires a higher
amount of binder, that
increases the density of the boards. Therefore, the mechanical,
thermal and acoustical
properties significantly differ from the low density WWCB which
are covered by a very
thin layer of binder having a solid content by volume of <
30%.
Figure 1.2: Visualization of wood types. From left to right with
reduced dimensions: logs, lumber, veneer, wood chips, wood strands,
wood granules, wood slivers, wood shavings, wood flakes, wood wool,
wood fiber, wood dust, paper fiber, wood flour, and cellulose.
1.2 Raw materials
1.2.1 Cement
The use of Ordinary Portland Cement (PC) within WWCB provides
many advantages,
like high mechanical strength and long service life. This
property is due to a densification
of the matrix around the wood wool strands, known as
mineralization (Semple and
Evans, 2004) occurring during the aging of the boards in an
external environment. The
further formation of hydration products within the core of the
cellulose fibers, and
perhaps also in the fiber cell wall, leads to an increment of
strength and stiffness of the
panel (Bentur, 1989). Besides this, the mineralization allows
outdoor applications where
the weathering of the boards only causes small dimensional
changes (Simatupang and
Geimer, 1990).
-
10 Introduction
Meanwhile, the manufacture of WWCB also involves some negative
aspects as the PC
production generates around 1 ton of CO2 for each ton of cement
(Provis and Bernal,
2014). During transport and production also fossil fuels are
consumed and nitrogen
oxides (NOx) and sulfur oxides (SOx) emissions are emitted,
contributing to the
greenhouse effect (Graham, 2003). Moreover, for many years, the
economy operated as
an open system, where raw materials were continuously consumed,
returning in unused
by-products in form of waste. To control the waste stream,
effort is required in recycling
and re-using of waste and only then accumulation of the
environmental impact will start
to reduce (Ehrenfeld and Gertler, 1997). Therefore, in recent
years, many studies have
been focusing on reusing and treating waste products (now seen
as by-products) to be
applied in existing and new materials (Wolfe, 1999; Garca et
al., 2008; Cheah and Ramli,
2011; Ashori et al., 2012; Nazari et al., 2014; Tang et al.,
2015; Wong et al., 2015).
Besides the environmental issues, the compatibility between
cement and wood is still a
studied matter. Soluble extractives of the wood wool strands
could retard, even terminate,
the hardening process of the cement. This phenomenon could be
easily observed by
monitoring the heat release over time during the exothermic
hydration of the cement.
Simatupang (1992) shows that the absence of peaks corresponding
to the consumption of
specific phase in the cement (C3S and C3A) or their shift is an
evidence of this
phenomenon. This inhibitory effect of wood on cement is mainly
due to substances such
as hemicelluloses, starches, sugars, contained in the wood
fibers, which dissolve in the
high alkaline environment (Fan et al., 2012). Therefore, the
wood is stored and treated by
e.g. soaking of the wood, treatment with Ca(OH)2 and
acceleration of the hydration of
cement by additives like alkalis (Hachmi et al., 1990;
Simatupang and Geimer, 1990;
Jorge et al., 2004; Pereira et al., 2006; Fan et al., 2012) and
chlorides or even pre-
hydrated C3S (Young, 1972) e.g. from the recycled concrete fines
(Florea et al., 2014).
However, some of these solutions are not always applicable, e.g.
to materials like
WWCB the use of chlorides is restricted by the EN 13168 to a
very low quantity (
0.06% in case of Cl3 level). This is because often the boards
are assembled near
reinforced concrete structures in which penetration of chlorides
can cause erosion of the
reinforcement and damage the structure (Lee and Short, 1989;
Mencnarowski, 2014).
Ordinary Portland Cement is the second most-used commodities
worldwide after water
(Brouwers, 2010). Due to the high environmental impact many
studies are performed on
this topic. Among them, Damtoft et al., (2008) defines an
optimization of the PC
manufacture for the minimization of its CO2 footprint. The
application of bio fuel and
alternative raw materials for the production of the binder such
as limestone powder (LP),
ground granulated blast furnace slag (GGBFS) and fly ash (FA)
can lead to a maximum
reduction in the CO2 emission of 17% (Damtoft et al., 2008).
Despite of those
achievements, the PC carbon footprint needs to be further
reduced. Among the potential
alternatives, alkali activated binders (AAB) are very promising
and can lead in case of
concrete to a much lower environmental impact compared to the
use of Ordinary Portland
Cement.
The main difference between AAB and cement production is the
avoidance of a high
temperature calcination step, during the synthesis of FA or
GGBFS. Although the
minimal environmental impact of the production of those prime
materials (Provis, 2014),
-
Chapter 1 11
the use of hydroxide or silicate activation does increase the
greenhouse effect (Duxson et
al., 2007). Based on the carbon dioxide equivalence (CO2-e), an
evaluation of the CO2
footprint of PC and some alkali activators can be done. CO2-e is
a term for describing
different greenhouse gasses in a common unit. For any quantity
and greenhouse gas type,
CO2-e signifies the amount of CO2 that will have the same global
warming potential
(GWP) (Brander and Davis, 2012). Therefore, calculations of
emitted CO2-e (kg CO2-
e/kg material) is based on the sum of the contribution belonging
to CO2, CH4, NO2 and
synthetic gasses emissions, developed during a specific activity
and taking into account
the embodied energy of the material. NaOH carbon footprint is
mainly depending on its
manufacture. It is produced with chlorine through the alkali
process (processing alkalis
with electrolysis). On the other hand, Na2SiO3 is produced by
melting silica and sodium
carbonate (Turner and Collins, 2013). Table 1.1 displays the
estimated CO2-e value
corresponding to the environmental impact of their manufacture.
In case of white cement,
the lower amount of C4AF in the clinker requires a higher
temperature in the kiln, hence,
the CO2 footprint will be even higher than the one reported for
PC.
Table 1.1: CO2 footprint of fly ash (FA), PC, NaOH and Na2SiO3
(Turner and Collins, 2013).
FA PC NaOH Na2SiO3
kg CO2-e/kg 0.03 0.82 1.92 1.51
The values in Table 1.1 take into account the CO2-e for the
materials manufacture.
Although the alkali activators are characterized by high
environmental impact, the
quantity necessary for the activation of AAB will be much less
compared to the use of PC.
In case of AAB the real saving lies in the use of sustainable
raw materials, as FA and
GGBFS. The utilization of these prime materials will provide
improvements connected to
the removal of materials from landfill, and an alternative
utilization of the waste streams
(Duxson et al., 2007). Further developments of this technology
can lead to the application
of binding systems based on the partial or total replacement of
PC, reducing to the
minimum the PC content and also its environmental impact.
1.2.2 Wood
In Europe, mostly coniferous species (softwood), like spruce,
are used to produce
WWCB due to their high availability and lower degradation by the
alkaline environment.
This leads to a lower release of inhibitors compared to
deciduous species (hardwood)
(Beltran Sierra, 2011). Also the use of aspen, basswood,
cottonwood, black willow and
yellow-poplar is mentioned (U.S. Department of Agriculture,
1987). Lee and Short,
(1989) concluded that WWCB can be produced by using hardwood
species like southern
pine and confirmed the use of cotton wood when pre-treating with
3% CaCl2, 3%
Na2SiO3 or plain water. Yellow polar could be applied when
treated with CaCl2 or
Na2SiO3 and sweetgum when using CaCl2. Some other suitable
wood-species are
presented in Table 1.2, making it clear that most wood species
around the globe are
feasible for production of WWCB, indicating that WWCB can be
produced everywhere
in the world (van Elten, 2015).
In this study, Spruce wood is mainly used and represents the
largest part of the standing
tree species in Sweden with approximately 44% and is one of the
most commonly used
wood species in the Northern Europe (Brndstrm, 2001a). Spruce
consists of many
-
12 Introduction
pores having a hexagonal shape as presented in Figure 1.3.
Because of the high amount
of voids, the wood wool strands have a porosity up to 63%. Due
to this porosity, the
thermal conductivity of oven dry spruce is 0.11 W/mK (Simpson
and Tenwolde, 1999;
Niemz et al., 2010) that makes it suitable for lightweight
thermal insulating panels. The
thickness of the cell wall and amount of cell walls are
determining the total porosity of
the spruce wood and vary per season, e.g. in spring cell walls
are thicker (2-3 m) than
cell walls in fall (+/- 7 m) (Proek et al., 2015). Furthermore,
the cells are denser packed
during fall compared to spring.
Table 1.2: Wood types successfully applied for WWCB production
(van Elten, 2015).
America Australia Europe Africa Asia
Alophyllum
brasiliense Acacia saligna Abies pectinata Alstonia boonei
Haldina
Cedrela Casuarina spp Abies Alba Androstachys Agathis
dammara
Ceiba pentandra Excoecaria agallocha
Castanea sativa Anigre Altingia excelsa
Pinus caribaea Picea Abies Aucoumea klaineana Neolamarckia
cadamba Picea Sitchensis Pinus Sylvestris Berlinia grandiflora
Artocarpus chaplasha
Brachystegia leonensis Bombax ceiba
Tiama Entandrophragma angolense
Toona ciliata
Entandrophragma utile Dipterocarpus alatus
Dipterocarpus gracilis Dipterocarpus turbinatus
Dipterocarpus sp. Eucalyptus deglupta
Figure 1.3: Scan of spruce wood wool strands (Doudart de la Gre
et al., 2013).
Wood structure
The wood pores as visualized in Figure 1.4 showing the open
parts of tracheids.
Tracheids, which comprise over 90% of the total cell wall
volume, serve as mechanical
support as well as conduction of water (Brndstrm, 2001). The
voids within a tracheid
are called lumen. Tracheids consist of single elongated cells
with lengths between
2-4 mm, a diameter of 20-40 m, wall thicknesses of 2-10 m and
are arranged
longitudinally (e.g. parallel to the growth of the trunk). The
tracheid cell walls are
constructed in a composite structure of a primary and a
secondary wall. The primary wall
and secondary wall consist mainly of cellulose and lignin (Gindl
et al., 2004). The fibrils
are composed by a practically equal amount of hemicellulose and
polysaccharides. This
lignin is an important factor in the determination of the
strength of wood because it is
binding the individual cells together (Gindl et al., 2004).
-
Chapter 1 13
(a)
(b)
Figure 1.4: (a). Diagrammatic view of the cell wall of a typical
coniferous tracheid. P primary wall; M
middle lamella; S1 outer layer of the secondary wall; S2 middle
layer of the secondary wall; S3 inner layer of the secondary wall;
W warty layer; P' and P'' primary walls of adjoining cells; (b)
Shows the difference between early thin-walled tracheids and late
thick-walled tracheids (Zimmermann, 1983).
Studies have indicated a difference in lignin content between
wood earlywood, harvested
in an early stage of the year (e.g. start of growing season) and
latewood, wood harvested
in a later stage (e.g. second half of growing season) (Proek et
al., 2015). Considering this,
the strength of wood is a result of the composite formed by the
interaction between
cellulose and lignin (approximately 25-33% of dry wood volume)
wound around. In
Table 2.2 of Chapter 2 the chemical composition of spruce wood
is provided.
Moisture content
Tracheids enable the transport of water through the pores
(Brndstrm, 2001). The
presence of water in softwood can be explained by the lumen in
the tracheids that serve
as water transfer, establishing a dissemination of water through
the whole trunk. One
month after harvesting the water content can decrease to
approximately 31% of its
original value (Laurila and Lauhanen, 2010). The moisture
content (Mc) of spruce can
reach values up to 53% directly after trunk harvesting (Laurila
and Lauhanen, 2010). Mc
can be determined according to
100i cell
cell
m mMc
m
(1.1)
where Mc is the moisture content of wood in [%]; mi is the mass
of water impregnated
wood in [g] and mcell is the mass of wood after 24h oven drying
in [g]. After oven drying,
wood wool strands absorb moisture naturally from the environment
although temperature
and relative humidity depending as illustrated in Figure 1.5
using the Hailwood and
Horrobin (1946) equation for wood.
-
14 Introduction
Figure 1.5: Moisture content absorption based on relative
humidity (Hailwood and Horrobin 1946).
As for the Mc, moisture can be located in the cell walls until
30% of its dry-oven mass
(saturation point of wood) and subsequently in the volume of the
lumen. When the cell
walls are completely filled with water, the maximum expansion of
the cell wall volume is
reached. When observing wood, the total volume of wood can be
expressed as the
volume of the wood cell walls ( cellV ) (expressed as a solid)
and the volume of the voids
of the wood ( voidsV ). The additional volume which is created
by the expansion of the cell
walls due to the absorbed moisture ( voidV ) together with the
volume of the lumen ( lumenV ) can be expressed as the total void
volume of the wood ( voidsV ) as presented in Figure 1.6.
The approach of having a fixed volume of wood but varying voidsV
differs from Siau (1984) who defines a total volume change, and by
that a varying density of the cell wall
with the Mc of the wood. The derivation of the known equations
following the
fundamental consideration is described on the next pages.
(a) (c)
Figure 1.6: (a) SEM picture of the cross section of a cement
covered wood wool strand; (b) Enlargement of
the cell walls and lumen, the cell wall structures are composed
of an inner, middle and outer layer, a primary wall and middle
lamella; (c) schematic representation of the total volume of wood
that consists of
voidsV and cellV of which voidsV is composed of voidV and lumenV
.
(b)
-
Chapter 1 15
Because the cell walls swell and shrink in the range of 0-30%
Mc, voidsV increases or decreases, respectively but remains stable
when the Mc is higher than 30% based on
oven-dry mass. In case the Mc has reached the saturation point
or higher, the volume of
the cell walls stays constant and the volume of the wood is
known as the green volume of
wood.
Using the approach of having a fixed volume of wood the maximum
Mc ( maxMc ) can be calculated when voidsV is completely filled
with water:
max 100w voids
cell
VMc
m
(1.2)
In which maxMc is the maximum Mc of the wood [%]; w is the
density of water in [g/cm3]
and Vvoids the void volume of the wood [cm3]. Furthermore, a
common expression used in
the wood industry is the specific gravity of wood (determined by
measuring the oven-dry
mass of the wood divided by the green volume of the wood)
divided by the specific
density of water following:
cell
cell voidsb
w
m
V VG
(1.3)
In which bG is the dimensionless specific gravity of wood
(measured by using the green
volume of wood) [-] and cellV the volume of the cell walls
[cm3]. Eq. (1.3) can be rewritten as:
cell
voids cell
b w
mV V
G
(1.4)
Substituting eq. (1.4) into eq. (1.5) yields:
max 100
cellw cell
b w
cell
mV
GMc
m
(1.5)
mcell can be rewritten as:
cell cell cellm V (1.6)
In which cell is the specific density of the cell walls in
g/cm3.
Substituting eq. (1.5) and (1.6) into (1.7) yields:
max 100
cell cellw cell
b w
cell cell
VV
GMc
V
(1.7)
-
16 Introduction
Substituting the value of water (1 g/cm3), Eq. (6) can be
rewritten as eq. (1.8) by:
max( )
100cell b
cell b
GMc
G
(1.8)
According to the U.S. Department of Agriculture (1987) the
dimensionless specific
gravity of spruce wood, bG is 0.36. Based on the specific
density of cell walls cell is
1.54 g/cm3, maxMc according to eq. (1.8) is 212%. Practically
such high Mc does not easily occur, since wood logs will not be
fully saturated with water. However, in case of
wood wool strands, this condition occurs easier due to the small
dimensions of the wood
wool strands. The ability of wood to absorb high amounts of
water, makes the production
more challenging since different cement types require certain
water amounts to react and
in the above mentioned production process sprinkling wood wool
with cement requires
cement to take water from the wood wool. Insufficient water will
therefore result in
insufficient hardening of the cement while in case excessive
moisture is present in the
lumen of the wood wool strands, this moisture will be liberated
during the press curing of
the composite and cement paste migrates to the bottom of the
boards.
1.3 Production process WWCB in the Netherlands
Since the method of manufacturing a material influences the
final properties of the
produced materials, the production process needs to be well
understood to ensure low
deviations in the products performance. The first impression
that people have when
introducing WWCB is that it is a low tech material. In reality
many factors are at hand
which gives the material its unique features. In the following
sections, different facets of
the production processes to produce WWCB are described. In
short, the process is
divided into four main stages, i.e. conditioning of wood logs,
shredding of wood logs,
forming of the composites and curing of the composites. The
influences of several
parameters on the above stated stages of the production process
are afterwards briefly
discussed and analyzed in relation to the most important
properties of the boards. Based
on the obtained findings, suggestions for improvements are
provided. Figure 1.7 shows a
visualization of a wood-log, wood wool strands and a WWCB.
Wood log Wood wool strands WWCB (a) (b) (c)
Figure 1.7: Visualization of the (a) Wood logs; (b) Wood wool
strands; (c) WWCB. Pictures made by
Botterman (2016).
Forest trees, preferably PEFC or FSC certified, such as pine,
spruce, poplar or eucalyptus
(van Elten, 2006), are harvested at 30-50 cm above the ground
level, whereas multiples
of 0.5 m are used for the production of WWCB. The diameter of
the logs should satisfy
the required diameter range of 16-26 cm. The branches and bark
of the trees are removed
on site. After sawing, the wood logs, known as green wood, are
stored on site for 3-6
-
Chapter 1 17
months in order to leach the soluble sugars out and reduce the
moisture content (Mc), also
known as wood seasoning. The wood logs and wood wool strands
further described in
this study are all from Norwegian spruce (or simply termed
spruce wood). Initially the
Mc depends on the age of the wood log, type of wood, seasoning
of the wood log,
geographic origin etc. The measurements of 8 green and 8
seasoned spruce wood logs in
February 2015, based on mass of oven-dry wood indicate a Mc of
63-102% and 22-63%,
respectively. The chemical analysis of green and seasoned spruce
wood showed only
minor differences and are likely caused by its natural deviation
rather than the influence
of seasoning. The production process of WWCB starting from the
seasoning of wood
logs follows the procedure, as presented in Figure 1.8.
Figure 1.8: Modified schematic overview of the production
process at Knauf Insulation Oosterhout (Mencnarowski, 2014).
First, the outdoor stored wood logs are cut into blocks of 50 cm
in length, screened for
metal parts and cut into 25 cm pieces and shredded to wood wool
strands. Although there
are still machines that produce longer strands, the length of
the wood wool strands is
reduced by the introduction of the Eltomatic Rotating Wood Wool
Machine also named
CVS-16 in which CVS stands for continuously variable speed and
16 the amount of slot
knives (van Elten, 2015). The final dimensions of the wood wool
strands are 25 cm in
length, 1-3 mm in width and 0.15-0.4 mm in thickness. The Mc of
the produced spruce
wood wool strands varies between 13-27% (determined on 17
randomly taken samples
during the production of wood wool strands in June and July
2014, at Knauf insulation).
The Mc of the wood wool strands is always lower than the initial
Mc of the outside stored
wood logs that enter the cutting machine because part of the
moisture is evaporated
during the shredding process in which fast rotating (high
temperature) knives are used.
The variation of the Mc within the wood wool strands can be
attributed to the nature of
the wood, as indicated before, but also to the storage
conditions exposing the wood logs
to outdoor climate conditions as presented in Figure 1.9. During
dry days water
evaporates while after rainy days, the measured Mc in the wood
wool strands is
increased. Still on average (Mc ~18%) the moisture content is
lower in June and July
-
18 Introduction
(summer period) compared to the earlier reported moisture
content in February (winter
period).
Figure 1.9: Weather conditions in June and July (data from
weather station Oosterhout the Netherlands) and Mc of the wood wool
strands (based on oven-dry mass). The bars indicate rain
periods.
After shredding, the wood wool strands are dipped in a water
bath to increase the Mc,
which is later needed for the cement hydration. Due to the high
water uptake of the wood
wool strands the wood wool strands are pressed with a roll press
to decrease the water
content. The opening of the press is adjustable depending on the
water content of the
wood wool strands.
The most influential factors regarding the water uptake are the
initial Mc of the wood
wool strands, the dimensions of the wood wool strands (surface
area), the density of the
wood wool strands (cell walls that chemically can bond water and
pores that can capillary
absorb water) and the opening of the roll press (Figure 1.10a).
Moreover, Na2SiO3
(sodium silicate) can be added into the water bath to enhance
the cement reaction by the
availability of dissolved Si ions which react with the Ca ions
from the cement to produce
C-S-H gel and elaborated heat. Afterwards, the wet wood wool
strands, together with
cement powder, are fed into a rotating mixer. The amount of
cement is depending on (1)
the amount of wet-wood wool; (2) the mass of the wet wood wool
strands on the belt and
(3) the mass variations of the belt (Figure 1.10b).
The mixing time is related to the amount of wet-wood wool that
is fed to the mixer and
the amount of cement covering the wood wool strands. The
irregular flow of wet wood
wool strands is continuously controlled by an electronic device.
The double distribution
machines spread two different layers of a continuous mat of wood
wool strands cement
-
Chapter 1 19
on top of each other into the molds. Based on the type of WWCB,
e.g. made of 1.0, 1.5,
2.0 or even 3.0 mm wide wood wool strands with various board
thicknesses, the
distribution machine allows higher or lower quantities of
mineralized wood wool strands,
controlled by the program.
(a) (b)
Figure 1.10: (a) Pressing belt to reduce moisture content of the
wood wool strands after soaking in a water bath; (b) Weight
measurements of wet wood wool strands and belt.
After having passed a hydraulic pre-press roll (small force,
enough to press the composite
together), the molds are separated by a circular saw and moved
to the hydraulic stacking
press. This machine stacks the molds with fresh material (the
mold height is used as a
reference for the board thickness). When the maximum stack
height is reached, the stack
is moved out and stored under pressure again for 24 h (e.g. by a
concrete block of 1500
kg). After the setting of cement, the boards are taken from the
molds for further curing,
while the molds can be cleaned and oiled for re-use.
After a storage period of 10 days, the boards are transferred
into an oven at 130-160 C
for at least 30 min to remove the excess water, to limit the
shrinkage of the boards and to
reduce the unit weight. The Mc of the boards after heating is
around 12% by dry mass of
WWCB. Finally, the boards are painted, stacked and packed.
1.3 Influential parameters and consequences
Shredding the wood wool strands (geometry)
The wood logs (25 cm in length) are shaved in the CVS-16 by
being pressed against the
knives which are placed in a rotating disk (Figure 1.11a), and
by this an external load is
created on both the knives and wood logs. This load would be
higher when applying long
wood logs, hence, the length is restricted. Depending on the
aesthetic and func tional
requirements of the boards, wood wool strands with different
dimensions can be
produced with a width of 1-3 mm and a thickness of 0.15-0.5 mm.
The width depends on
the distance of the grooves of the knives (van Elten, 2010)
(Figure 1.11b). The thickness
of the wood wool strands depends on four different factors: (1)
the speed ratio between
the rotating knives and the feeding speed of the wood logs; (2)
the wearing/abrasion of
the knives, e.g. after a certain production time the knives need
to be replaced since the
thickness of the strands decreases; (3) the force used to press
the wood logs against the
-
20 Introduction
knives; (4) the Mc and density of the wood logs. Regarding the
latter, fresh logs contain
high amounts of moisture, and when the content is too high (>
35% based on dry mass),
short and thin wood wool strands are produced with a thickness
of 0.240 mm compared
to 0.340 mm when using dry wood-logs (Mc < 30%). This is
caused by the increased
resistance of the wood-logs against compression when the pores
in the wood wool strands
are filled with liquid water, hence, the knives will penetrate
less deep into the wood-log
and no proper wood wool strands are produced (van Elten, 2015).
In the case of logs with
a high Mc, the feed speed through the knives can be increased
(up to a certain level) to
create thicker wood wool strands. However, this results in a
higher load on the knives
that leads to a reduced life time and increased energy
consumption of the machine.
Furthermore, the wet wood wool strands clump easily together
with the dust and
eventually lead to the obstruction of the output passage of the
machine. In overall, the
thickness of the wood strands are difficult to control, and
highly depend on the used
machine settings and the used wood logs. In case of wood logs
with a low Mc (< 20%),
there will be more friction (measured energy consumption is 10%
higher compared to
wood logs with a Mc of 30%) that leads to an increased
temperature of the knives and
local combustion of the wood wool strands. Therefore, it is
recommended to keep the Mc
between 20-35% on dry mass of the wood, meaning that the cell
walls of the wood are
almost or completely saturated but the location of liquid in the
lumen of the wood is
limited. In this way the friction between the wood logs and the
knives is reduced leading
to a lower energy consumption of the knives and prolonged
lifetime (Figure 1.11c).
(a) (b) (c)
Figure 1.11: (a) Rotating disk with 16 knife slots; (b) Example
of a knife; (c) Schematic overview of the energy consumption of the
knives during the production process for a fluctuating and stable
log input.
Figure 1.12 presents the apparent density of 370 wood logs. The
observed density varies
between 400 and 700 kg/m3, fluctuating moisture contents and the
density of the log
(amount of cell wall structures present in the wood). When
monitoring the energy
consumption in time to shredder the wood logs, the energy
consumption varied between
35-54% in which 100% equals 160 kW with 900 rpm. Therefore, a
test is performed
where the wood-logs were separated into three different density
classes of which the
highest density class between 580-680 kg/m3 is shredded and
presented in Figure 1.13. It
is found that the energy consumption by shredding this high
density class fluctuates
around 51% with a deviation of 1%. Based on the energy
consumption of the high density
class it is recommended to separate them in density classes and
store the wood logs with
higher density classes longer to reduce the moisture content. In
this way the energy
consumption can be reduced and the life time of the knives can
be prolonged. Another
method would be kiln seasoning the wood logs. Although this
requires a heat source.
-
Chapter 1 21
Figure 1.12: Measured apparent density of spruce wood logs at
Knauf insulation Oosterhout.
Figure 1.13: Percentage of energy consumption required for
shredding wood logs with a density of 580-
680 kg/m3 in which 100% equals 160 kW.
Soaking of the wood
Because during the industrial production of WWCB the amount of
cement applied is
depending on the wet wood wool entering the mixer, the moisture
content of the wood
wool can have significant impact on the final composition of the
board e.g. wood, water
and cement amount. For example, in the first case there is a
board produced with 1 m3 of
wood wool strand having a Mc of 230%, and 1100 kg of binder. In
the second case there
is a wood wool strand volume of 1.2 m3 used with a lower water
content e.g. 190% and
again 1100 kg of binder. The properties of both boards will be
different. This since in
case two the same binder amount need to cover more wood wool
surface. Moreover, the
hardening of the board will be negatively influenced due to the
lower amount of moisture
available for the cement hydration.
-
22 Introduction
The moisture content of the wood wool strands is depending on
soaking of the wood
wool strands in a water bath and subsequently removing of the
water by pressing in a role
press. The fluctuating amount of wood wool strands through the
roll press (the wood
wool is not homogenously distributed over the belt) leads to a
fluctuating distance
between the cylinders of the role press, hence, the amount of
water within the wood wool
strands is deviating. The measurements of 17 randomly taken
samples of wood wool
strands after the water bath and role press indicate a Mc
varying between 182% and
295% (Mencnarowski, 2014).
Therefore, a study is performed to investigate the possibility
of characterizing the
influence of moisture on the total wood amount by reproducing
several board
compositions of already produced boards that are cured for 10
days. The moisture amount
is measured by drying the boards in an oven at 100 degrees for
24 hours and calculate the
mass loss. The wood amount is determined by measuring the loss
on ignition at 750
degrees for 1 h (Quiroga and Rintoul, 2015). The remaining mass
amount is accounted as
unreacted binder. Note that the values are only used for
indicative purposes, this since
there is a small percentage of unburned wood wool particles,
which are accounted as
binder. Moreover, there is an amount of chemical bound water
that is released by
dehydration of the binder and is accounted to the wood. To
define the latter is found to be
difficult since the hydration degree of the binder is varying
because of the moisture
uptake of the wood wool. The results of the study in which the
binder amount is linked to
the total wood wool and moisture amount is presented in Figure
1.14a revealing a clear
relation which is a result of the designed input recipe (the
fixed wood + moisture to
binder ratio), which controls the dosing unit of the binder. In
this case the moisture
amount presents the water contained by wood (measured by oven
drying) plus added
water for cement hydration. In Figure 1.14b the obtained wood
amount and moisture
amount are presented. It demonstrates there is no clear relation
between the moisture
amount and wood amount, which is caused by a number of factors
including the wood
density, moisture uptake, and the capillary water present in the
voids of the hardened
binder.
Nevertheless, in this production process, ideally, the wood wool
strands should first be
acclimatized to reach for instance the fiber saturation point to
obtain a constant mass. An
additional amount of water can then be added, taking into
account the water amount
required for the binder to hydrate, and the amount of water that
the wood wool strands
will keep due to chemical and capillary forces.
-
Chapter 1 23
(a) (b)
Figure 1.14: (a) Wood + moisture to binder ratio in mass; (b)
Wood to moisture ratio in mass.
1.4 The influence of moisture and binder amount
In addition to the previous studies, a study is performed on the
bending and compressive
strengths of boards produced with a higher (1.5 HB) and lower
amount of binder (1.5 LB).
The results of this study are presented in Figure 1.15. Due to
confidential reasons, the
amounts (recipes) are not displayed and the performed study is
only indicative. Initially
the aim of producing boards with a higher binder amount is to
better cover the wood wool
strands and by that to increase the fire resistant properties of
the boards. Moreover, the
hypotheses that boards made with more binder lead to higher
mechanical properties can
be answered. Figure 1.15 indicates that when observing the
compressive and bending
strength results it is difficult to distinguish between lower
and higher binder amounts.
This is proved to be caused by an improper reaction of the
binder probably related to the
applied water amount. Therefore, additional tests were
conducted, in which the water
amount was designed to assure a full reaction of the binder. The
results of this treatment
led to boards with bending and compressive strength properties
presented in Figure 1.16.
The results show a clear separation in density of the increased
binder amount. The
strength properties show that the increased binder amount leads
to a higher compressive
strength, but does not significantly contribute to an increased
bending strength. Therefore,
an evaluation of the stress and strain curves of the tested
boards is performed of which
some curves are presented in Figure 1.17. It is revealed that
the behavior of the composite
is altered by the modification of the binder amount. By the
increased binder amount the
strain reduces since the wood wool strands become less flexible
and will therefore
contribute less to the composite properties. Hence, with the
higher binder amount the
applied load can be higher before a board breaks. However, a
rapid failure of the test
specimens occur when reaching the breaking point, while with a
lower binder amount the
maximum load is lower and the boards first deform.
-
24 Introduction
(a) (b)
Figure 1.15: (a) Compressive strength and (b) Bending strength
of 1.5 mm WWCB produced with higher
binder amounts (HB) and lower binder amounts (LB).
(a) (b)
Figure 1.16: (a) Compressive strength and (b) Bending strength
of water-pretreated 2.0 mm WWCB with higher binder amounts (HB) and
lower binder amounts (LB).
(a) (b)
Figure 1.17: Stress strain curves of (a) Compressive and (b)
Bending strength of water-pretreated 2.0 mm WWCB with higher binder
amounts (HB) and lower binder amounts (LB) and a
water-pretreatment.
-
Chapter 1 25
1.5 Developments in WWCB
So far only boards are introduced, with consisting of wood wool
strands covered with
cement. However, since the start of WWCB production, the
industry is seeking for
utilization opportunities and the market needs introducing
additional products like
reinforced WWCB, sandwich boards and large WWCB elements.
Reinforced WWCB consist of 3 rectangular hardwood slates of 20 x
20 mm2 placed at a
distance of 150 mm in a board of 600 mm. The slates increase the
bending strength
providing an additional load bearing capacity and can be applied
on ceilings in garages
and as roof covering. The benefit of this material is its high
thermal capacity that reduces
the fluctuations of room temperatures in summer time. To fulfill
the demand of thermal
insulation, sandwich boards are introduced (Figure 1.18)
consisting of an WWCB layer
on which PIR, EPS, XPS, mineral wool or PF (resole foam) can be
attached.
Figure 1.18: WWCB ingredients and boards combined with EPS and
mineral wool. Illustration taken from www.eltomation.com with
approval of van Elten (2006).
Nowadays, the production of a wood wool strands cement walls
with a thickness up to
0.4 m has been successfully executed (van Elten, 2008). This
large WWCB element has
dimensions of 2.8 x 6.0 x 0.4 m3 (H x W x D) with a density of
330-350 kg/m3 (van Elten,
2015). This type of product provides thermal insulation values
(Rc) of ~5 m2K/W with a
thermal conductivity () of 0.08 W/mK. The product is interesting
because it consists of
locally available raw ingredients (wood, cement and water)
(Ashori et al., 2011) and its
resistance against termites (Eusebio et al., 2000).
1.6 Conclusions
The production of wood wool strands cement boards is extensively
discussed in this
Chapter, giving an overview on the complexity of the boards and
their properties. During
production, many parameters are influencing the final product
like the storage conditions,
used knives, recipes and material dosage. Overall, moisture is
in all stages a very
influential parameter, which is practically difficult to control
in the current production
process. Moreover, the raw materials cement and wood are
addressed, indicating the
importance to lower the environmental footprint of the boards
and providing insights in
the basic wood structure. Based on the study, the following
conclusions can be drawn:
http://www.eltomation.com/
-
26 Introduction
The moisture content (Mc) of the wood logs together with the
density of the wood-logs in terms of dry-mass, have a significant
effect on the industrial
production of WWCB. They influence not only the life time of
cutting knives and
energy consumption of the wood wool strands shredding machines,
but also the
dimensions of the produced wood wool strands. It is recommended
that the Mc of
the wood-logs for board production is in de range of 20-35%
based on oven-dry
wood.
By dividing wood-logs in different density classes, and shredder
the low density wood-logs with a proper moisture content, the
lifetime of knives in the shredding
machine can be prolonged and the deviation in the dimensions of
wood wool
strands can be minimized.
The defined ratio between wood + moisture and binder amount is
found to be successfully utilized in the production process.
However, since the moisture and
wood amount in the parameter wood + moisture is fluctuating, no
clear relation
could be described between the three ingredients and the final
board properties.
This while the influence of a higher binder amount or lower wood
volume has
significant impact on the final board properties.
In case of applying a higher binder amount, application of
sufficient water is of high importance, to obtain boards with the
required properties.
Compatibility between wood and cement is still a studied matter,
as well as the use of alternative binders in combination with wood.
The use of alternative
binders is by its low environmental footprint worth
researching.
The total water demand required for WWCB during the industrial
production can be divided into: (1) the moisture to reach the
saturation point of the used wood
(30% of the dry oven mass); (2) the water demand that cement
particles can
absorb from the wood wool to fully react; the later depends on
the cement type.
1.7 Problem statement
Retardation effect
Although the background study reveals that wood cement products
are used for centuries,
the combination is not as straightforward because of the
interaction mechanism between
the wood and cement leading to a postponed hardening of cement
known as the
retardation effect. The mechanism of retardation on cement, was
extensively studied in
the 1970s and 1980s (Yamamoto, 1972; Young, 1972; Popovics,
1976; Milestone, 1977;
Thomas and Birchall, 1983), although still not fully understood.
Nevertheless, a common
solution to prevent retardation is by using accelerators like
Na2SiO3, CaCl2 and MgCl2 (Hachmi et al., 1990; Simatupang and
Geimer, 1990; Jorge et al., 2004; Pereira et al.,
2006; Fan et al., 2012). However, the use of chlorides for WWCB
is restricted by
EN 13168 in which specifications are provided regarding factory
made wood wool
products.
Environmental footprint
Despite the advantageous application of cement in
lignocellulosic composites, the
increasing worldwide awareness of the substantial contribution
of Portland cement (PC)
-
Chapter 1 27
to greenhouse gas emissions (Florea et al., 2014;
Ramezanianpour, 2014) results in
searching for alternative materials and development of new
binders (Pacheco-Torgal et
al., 2008, 2012; Provis, 2014). The design of lignocellulose
composites with alternative
binders potentially have environmental and economic advantages
e.g. reduction of CO 2
emissions and raw materials that go to landfill as well as lower
costs (van Elten, 2015).
As of today, supplementary materials like limestone powder, fly
ash or ground granulated
blast furnace slag are commonly used as cement replacement.
Nevertheless, studies also
reveal that higher substitution levels drastically influence the
early strength development
of the binders. Therefore, the addition of alkalis to enhance
the dissolution of the
supplementary materials has been introduced that allows full
replacement of cement
(Yang et al., 2013; Ouellet-Plamondon and Habert, 2015).
However, the question remains
if alkali activated binders can be used with wood wool since the
exposure of
lignocellulose in an alkaline environment initiates several
reactive phenomena, e.g.
dissolution of organic acids, polysaccharides, peeling
reactions, hydrolysis of glycosidic
bonds and acyl groups (Mirahmadi et al., 2010).
Board properties
Depending on the used binder, the chemical composition and
internal structure of the
reaction products differ. Therefore, the material properties of
the composite can vary
distinctively when different types of binders are used
(Reinhardt, 1998; Pacheco-Torgal
et al., 2008; Ramezanianpour, 2014). Moreover, due to new
production techniques and
material innovations, competitive materials are rapidly entering
the market while the
acceptance of a material is defined by its ability to satisfy
multiple requirements. A
material with a range of contributions results in more suitable
product for various
conditions and special applications. To stay competitive, clear
insight in WWCB
properties is needed that allows a tailor-made WWCB.
1.8 Research targets
Objectives and methodology
The research is carried out by performing a systematic study to
address the above stated
challenges and provides solutions that can be directly applied
in practice. This study is
performed in close cooperation with the companies Knauf
Insulation, Eltomation, ENCI
and Van Gansewinkel Minerals to provide information which is not
addressed in
literature. Overall, this thesis provides some insights in the
influential parameters during
production, reaction mechanisms of applied binders,
microstructural adjustments that
alter the board properties and disposal and reuse of WWCB after
its lifetime. The thesis
consists of three core elements, whose objectives and
methodology are provided in the
following sections.
Part I: Binder
The objective of the first part of this thesis is twofold (1)
reducing the retardation effect
by sugars and (2) reduction of the environmental footprint of
wood wool composites
which both are connected to the applied binder.
-
28 Introduction
To investigate the retardation mechanism of wood on cement
hydration, a simplified
method by using glucose is applied.
The retarding effect on several cement phases is measured to
indicate to which extent
C3A and C4AF with varying anhydrite percentages is affected,
with the aim to reduce the
retardation on the calcium silicate hydrates. By using an
isothermal calorimeter with
different Portland cement types, clinkers and additions of
glucose, the heat release of the
mixtures is analyzed. The reaction product in the form of
portlandite is then further
analyzed at different hydration times by Thermal Gravimetric
Analysis (TGA) and the
microstructure was analyzed by Scanning Electron Microscope
(SEM). Finally, mortars
and wood cement composites are created to validate the present
findings.
The objective of the reduction of the environmental footprint is
divided into (i)
application of Papersludge fly ash (PsFA) to WWCB and
implementation of an optimized
mix design to determine the rate of cement replacement and
influence on the board
properties (ii), to evaluate the mechanism occurring when an
alkali activated binder is
used in the presence of wood, and to define the pre-treatment
needed to reduce the liquid
uptake by the hygroscopic behavior of wood.
For the application of PsFA, two treatment pathways are
evaluated to increase the
usability of supplementary cementitious materials (SCMS), in
this case, industrially
calcined PsFA by reducing its water demand. The PsFA, both
treated and untreated, are
physically and chemically characterized. By monitoring the heat
release, the reactivity of
the binder is examined. The suitability of PsFA as a cement
substitution is then evaluated,
by producing wood wool composites and defining the influence on
the properties such as
thermal insulation and sound absorption.
Furthermore, an eco-friendly binder is designed with not a
porous matrix but a matrix
with lower amount of voids optimizing the theoretical particle
packing density of the
binder by using the modified Andreasen and Andersen model.
For the alkali activated binder, the wood behavior under
alkaline conditions is first
evaluated by pH measurements together with the reaction
mechanism of the studied
binders using an isothermal calorimeter. Next, the ionic
behavior of Ca 2+ and Na+ in the
absence and presence of wood is studied using an
ion-chromatography. Furthermore, the
location of liquid water using NMR in wood is studied. The
obtained knowledge led to
the design of a sustainable lignocellulosic mineral-binder
composite using different alkali
activated binders.
Part II: Development of board properties
The objectives of the second part of this thesis is twofold: (1)
to evaluate the suitability of
WWCB to be used as an air purifying material by implementing a
photocatalytic coating
and to determine the influence of the surface structure of the
board and (2) to model the
sound absorption of WWCB, by defining the required input
parameters and studying the
influence of WWCB parameters on the sound absorption.
To increase the functionality of the board by addressing the
indoor air purification of
WWCB, a photo-catalyst, titanium dioxide (TiO2), is applied on
different WWCB with
various surface structures and binders. Using a mixture of NO
and NO2 as the target
pollutants the degradation rate of these pollutants is
investigated under different airflows
to evaluate the influence of surface structure and binder type.
Moreover, a durability
study is conducted exposing boards to steady and dynamic water
treatments.
-
Chapter 1 29
To model the sound absorption of WWCB, first commercially
available WWCB are
measured using an impedance tube and the required input
parameters are obtained by
using the curve fitting approach. By using multiple samples with
a variety of densities
and using 1.0, 1.5 and 2.0 mm strand widths the most accurate
model is selected for
further use. The selected impedance model is then demonstrated
using different strand
widths and board thicknesses. Afterwards, the influence of board
ingredients and
properties like strand width, density, wood to binder ratio,
moisture content, board
thickness and application of an air cavity are studied together
followed by an
optimization study with extrapolating the obtained parameter
functions.
Part III: Disposition and Reuse of WWCB
The objective of the third part of this thesis is to investigate
the disposition of WWCB
after its service life following the waste hierarchy from re-use
to disposal at landfill sites.
Although this part is closely related to Part I, with the aim to
reduce the environmental
footprint, it focusses less on implementation of a binder and
board properties (Part II) and
more on the method to reuse minerals classified as unsuitable
for reuse (waste). It is
found that a common end step after the service life of WWCB is
energy recovery by
incineration. However, due to the large volume required for the
incineration and the
relative small waste volume of WWCB, the waste is often mixed
with other wood
products which are more contaminated, hence, it generates
contaminated fly ashes.
Therefore, four different fly ash types produced by two
different biomass incineration
plants were analysed and compared to Dutch and European
standards on building
materials. A combined treatment method was designed for lowering
the leaching of
contaminants and the effect of each treatment step was
quantified. Finally, a pilot test was
performed in order to scale up the treatment and make the
biomass fly ashes suitable to
be used as SCM.
1.9 Outline of this thesis
This research is based on multidisciplinary approaches and aims
to provide insight into
development of novel WWCB products that are environmentally
friendly and widely
applicable. Although WWCB is known since 1920, to the authors
knowledge, there are
only a handful of papers available on this topic. Knowledge of
application of alternative
binders in other materials, e.g. binders in concrete products,
is used in this study. Also,
application of widely used acoustic impedance models is applied
to WWCB to identify
the acoustical properties of boards with different physical
properties. Moreover, the
application of a photocatalyst and its influence on material air
purifying properties is
studied. The research framework of this thesis is presented in
Figure 1.19.
Chapter 1 is an introduction to WWCB, briefly addressing the
ingredients, the production
process and resulting board properties of commercial WWCB,
introducing influential
parameters and the complex interaction between the wood wool
strands and binder.
These results are obtained in the real scale production of WWCB.
After that, the core of
the thesis is organized into three parts, composed by
self-contained chapters:
Part I, composed by Chapter 2, Chapter 3 and Chapter 4,
describes different approaches to optimize the used binder and the
application of alternative binders
-
30 Introduction
to reduce the environmental footprint of the boards. Chapter 2
presents the study
on hydration of cements and the retardation of sugars on cements
with varying
aluminate and CaSO4 contents providing new insights to the
retardation
mechanism and a novel solution to reduce the retardation effect.
Chapter 3,
describes the implementation of PsFA as a supplementary material
leading to
increased mechanical and thermal properties and a reduced
environmental
footprint. In addition, densification of the binder matrix by
optimizing the particle
packing is evaluated. Chapter 4 describes the use of
alkali-activated binders, to
fully replace cement, identifying the alkaline degradation
mechanism of wood and
reaction mechanisms which have not been studied before for such
system
(mineral binder combined with wood wool). The study lead to the
development of
a hybrid binder that results in a carbon footprint reduction of
60% compared to
the use of cement as binder and leads to significant reduction
in costs.
Part II, composed by Chapter 5 and Chapter 6, describes
different approaches to optimize the board performances. Chapter 5,
describes a method to increase the
functionality of the boards by implementation of a
photocatalytic coating. The
study leads to increased insights in the surface influence and
ultimately in a high
efficiency of 98% (according to the knowledge of the author such
values have not
been reported before) in air purification using WWCB. Chapter 6,
describes an
approach to model the sound absorption of WWCB. Different
commercial
WWCB are characterized and, by using impedance models, the sound
absorption
of WWCB are predicted with varying strand widths, densities,
thicknesses and air
cavities providing new insights in the acoustic behavior of WWCB
and enabling
the enhancement of the sound absorbing properties by increased
densities and
multilayer constructions.
The content of the chapter is almost identical to the published
article (Botterman
et al., 2016). The experimental and modeling work is conducted
by Bram
Botterman and is part of his master thesis (Botterman, 2016).
During his study I
was his daily supervisor and we performed the research and
analyzes together.
Our collaboration of 1.5 years was followed by his graduation,
and we continued
to work together on writing the journal article (Botterman et
al., 2016), on which
this chapter is based on.
Part III, Chapter 7, considers the disposition of commercial
WWCB after its service life time and leaded to waste wood
incineration and contamination of
residues that required the design of economical feasible
treatments not seen
before. Moreover the designed treatment is evaluated on a pilot
scale, resulting in
a binder which could re-enter the market as a SCM.
In Chapter 8, comprehensive conclusions of the present work are
drawn and
recommendations for future research are provided.
-
Chapter 1 31
Chapter 1: Introduction
Chapter 2: Hydration kinetics of cement containing sugars
Chapter 3: Design of alternative binders in wood composites
Chapter 4: Ionic interaction and liquid absorption by wood in
lignocellulose inorganic mineral binder composites
Chapter 5: Advanced functionalities
Chapter 6: Modelling and optimization of the sound absorption of
wood-wool cement
boards
Chapter 7: Disposition after service life and treatment of
biomass fly ashes
Compositeproperties
Binder
Recycling
Chapter 8: Conclusions and recommendati